In multiple instance learning (MIL), instead of the instances, there are bags and each bag has certain number of instances. Given the bags with class labels, aim of MIL is to classify bags with potentially unlabelled instances. Some of the earlier MIL methods focus on solving MIL problem under the standard MIL assumption, which requires at least one positive instance in positive bags and all the remaining instances in the data are negative. Due to the restrictiveness of this assumption, generalized MIL assumptions are also introduced to increase applicability to various MIL problems.

MIL_datasets can be downloaded from here.

Click here to download the codes of the MIL algorihms.

This website can be referenced using the following bibtex entry:

@misc{milweb,
author = {Kucukasci, Emel S. and Baydogan, Mustafa G. },
title = {Multiple Instance Learning Repository},
url = {http://www.multipleinstancelearning.com/},
year = {2018}
}